Maternal diabetes (MD) increases risk for obesity and type 2 diabetes mellitus (T2DM) in adulthood. Our long-term goal is to understand offspring pancreatic B-cell damage by MD, and the adaptive mechanisms of the offspring B-cell to insulin resistance and what goes wrong in this adaptation during the pathogenesis of T2DM. Specifically we are interested in the role of free radical oxygen species (ROS) during pregnancy that may be critical for damaging offspring B-cell adaptation. The insulin secreting B-cell is one of the most sensitive cells in the body to ROS damage, and the levels of anti-ROS enzymes such as catalase, superoxide dismutase (SOD) are the lowest and are not up-regulated by hyperglycemia. Most of the support for the antioxidant hypothesis derives from treatment of diabetic rats with antioxidant vitamins, E and C. However, there is only a little evidence that comes from basic research on B-cell. In addition, most works were done on streptozotocin (STZ)-induced severe diabetic models that produce low birth weight offspring and more significant outcomes than that in mildly hyperglycemic models. In human, hyperglycemia in most diabetic women is under control before and during pregnancy, thus hyperglycemia is not severe, and the offspring birth weight of most diabetic mothers is high (macrosomic);therefore, STZ-induced severe MD is not like human MD. We have observed that severe MD induced aberrations in pancreatic islet. To test the ROS hypothesis and significance of offspring birth weight and the significance of severity of MD, we will apply our expertise in islet function to determine if insulin secretion is preserved by transgenic antioxidant proteins in the offspring of mild and severe diabetic mothers. We have produced fourteen lines of transgenic mice that overexpress catalase, manganese SOD (MnSOD) and metallothionein (MT) in B-cells, and we have tested that these transgenes have no deleterious effect on FVB mouse B-cell structure or function. We have showed that catalase, MnSOD and MT overexpression protects islets against STZ or ROS and found a high level of protection against diabetes. They provide an ideal model to test the hypothesis that ROS contribute to B-cell damage. We will produce two kinds of offspring from STZ-induced mild and severe diabetic mothers, and comparatively examine whether antioxidant treatment and these antioxidant transgenes in fetal B-cells can prevent or delay the development of T2DM in adulthood.